Differential Roles of Two Major Brain Structures,
Mushroom Bodies and Central Complex, for
Drosophila Male Courtship Behavior
Takaomi Sakai, Toshihiro Kitamoto
Department of Anesthesia, University of Iowa Carver College of Medicine, Iowa City, Iowa 52242
Received 12 August 2005; accepted 4 January 2006
ABSTRACT: Drosophila male courtship is a com-
plex and robust behavior, the potential for which is ge-
netically built into specific neural circuits in the central
nervous system. Previous studies using male-female
mosaics and the flies with defects in particular brain
structures implicated the critical central regions in-
volved in male courtship behavior. However, their acute
physiological roles in courtship regulation still largely
remain unknown. Using the temperature-sensitive
Dynamin mutation, shibirets1, here we demonstrate the
significance of two major brain structures, the mush-
room bodies and the central complex, in experience-in-
dependent aspects of male courtship. We show that
blocking of synaptic transmission in the mushroom body
intrinsic neurons significantly delays courtship initiation
and reduces the courtship activity by shortening the
courtship bout length when virgin females are used as a
sexual target. Interestingly, however, the same treat-
INTRODUCTION
Male courtship is arguably the most complex behav-
ior performed by the fruit fly Drosophila melano-
gaster. It is a stereotyped sequence of behavioral pat-
Correspondence to: T. Kitamoto (toshi-kitamoto@uiowa.edu).
Contract grant sponsor: Uehara Memorial Foundation (T.S.).
Contract grant sponsor: NIH; contract grant number: MH62684
(T.K.).
This article includes Supplementary Material available via the
Internet at http://www.interscience.wiley.com/jpages/0022-3034/
suppmat
' 2006 Wiley Periodicals, Inc.
Published online 3 May 2006 in Wiley InterScience (www.interscience.
wiley.com).
DOI 10.1002/neu.20262
ment affects neither initiation nor maintenance of court-
ship toward young males that release courtship-stimu-
lating pheromones different from those of virgin
females. In contrast, blocking of synaptic transmission
in a central complex substructure, the fan-shaped body,
slightly but significantly reduces courtship activity to-
ward both virgin females and young males with little
effect on courtship initiation. Taken together, our
results indicate that the neuronal activity in the mush-
room bodies plays an important role in responding to
female-specific sex pheromones that stimulate initiation
and maintenance of male courtship behavior, whereas
the fan-shaped body neurons are involved in mainte-
nance of male courtship regardless of the nature of
courtship-stimulating cues. ' 2006 Wiley Periodicals, Inc. J
Neurobiol 66: 821–834, 2006
Keywords: Drosophila melanogaster; courtship; mush-
room body; central complex; shibire
terns, composed of orientation, tapping, wing exten-
sion and vibration, licking, attempted copulation, and
copulation (Bastock and Manning, 1955; Cobb et al.,
1986; Welbergen et al., 1987). Multiple sensory
modalities, in particular chemical perception and
vision, play important roles in the regulation of dif-
ferent behavioral elements in male courtship (Hall,
1994; Greenspan and Ferveur, 2000). The main pher-
omone components of male courtship are cuticular
hydrocarbons (Jallon, 1984; Ferveur and Jallon,
1996), which are mainly nonvolatile and detected by
specific gustatory neurons (Bray and Amrein, 2003).
Volatile pheromones emitted by females also stimu-
late male courtship (Tompkins et al., 1980; Tomp-
kins, 1984). In addition, dynamic as well as static vis-
821
822 Sakai and Kitamoto

ual stimuli are needed for efficient courtship in D.

melanogaster (Markow and Manning, 1980; Tomp-
kins, 1984; Hall, 1994; Sakai et al., 1997, 2002).
Because a male fly kept in social isolation is fully
capable of carrying out the entire courtship in
response to the aphrodisiac sensory cues, the potential
for this behavior must be genetically built in the cen-
tral nervous system as the particular neural circuits
and the molecular machinery regulating their physiol-
ogy (Hotta and Benzer, 1976; Hall, 1979; Baker
et al., 2001). To understand how courtship behavior
is controlled by the nervous system, we need to iden-
tify the components of the relevant neural circuits in
the brain and elucidate how they interpret the multi-
ple sensory cues and regulate the highly coordinated
motor outputs.
The mushroom bodies (MBs) and the central com-
plex (CC) are two prominent structures in the Dro-
sophila brain [Fig. 1(A)] that have been extensively
studied in terms of their roles in a variety of behav-
iors. The MBs are paired structures composed of
thousands of intrinsic neurons called Kenyon cells
whose cell bodies are localized in the dorso-posterior
region of the brain. They receive multimodal sensory
information via the dendritic calyx mainly from the
olfactory centers of the antennal lobes (ALs) and
from other brain regions including the gustatory cen-
ters of the subesophageal ganglion (SOG) (Strausfeld
et al., 1998; Ito et al., 1998). The Kenyon cells send
axonal projections to the anterior part, where they
bifurcate to form the medial and vertical lobes. The
MBs are considered to be the ‘‘memory center’’, play-
ing critical roles in various learning and memory
paradigms such as olfactory conditioning, experi-
ence-dependent courtship suppression, and context
generalization in visual learning (Dubnau et al., Figure 1 GAL4 expression patterns in the adult brain of
2001; McGuire et al., 2001; McBride et al., 1999; Liu eight GAL4 lines. (A) Schematic representation of mush-
et al., 1999). In addition, nonlearning functions in room bodies (MBs), ellipsoid body (EB), and fan-shaped
walking (Martin et al., 1998) and centrophobism/thig- body (FB) in adult brain. (B–J) GAL4 expression patterns
motaxis (Besson and Martin, 2005) are also attributed in the brain of eight GAL4 lines visualized by GFP reporter
gene. F1 males between GAL4 females and UAS-GFP
to the MBs. Another significant structure is the CC,
males were used. We observed at least four males in each
which is located centrally in the brain. It consists of
line. (B) OK107. Arrow, pars intercerebralis (PI); arrow-
four characteristic neuropils, the protocerebral- heads, antennal lobes (ALs); asterisk, optic lobe. (C) 30Y.
bridge, the fan-shaped body [FB; Fig. 1(A)], the ellip- Arrow, subesophageal ganglion (SOG). (D) c772. (E,F)
soid body [EB; Fig. 1(A)], and the noduli. The CC J183. (G) OK348. (H) 104Y. Arrowhead, nerve bundle in
forms elaborate connections to a variety of brain lateral areas. (I) c232. (J) c41.
regions (Hanesch et al., 1989). Previous studies have
indicated the involvement of the CC in the olfactory
learning task (Heisenberg et al., 1985), as well as in ment of the CC in the CaM kinase-mediated memory
different features of motor outputs such as mainte- formation induced by courtship conditioning.
nance or temporal organization of locomotion Compared to the aforementioned functions of the
(Strauss and Heisenberg, 1993; Martin et al., 1999) MBs and the CC, their acute physiological roles in
and visual flight control (Ilius et al., 1994). Further- experience-independent aspects of male courtship
more, Joiner and Griffith (1999) revealed the involve- behavior are relatively unknown or still controversial.
Journal of Neurobiology. DOI 10.1002/neu

Previous studies reported that partial feminization of
the MBs by ectopically expressing the female form of
the sex-determining gene transformer (traF) resulted
in ‘‘bisexual’’ courtship, suggesting that mate dis-
crimination is dependent on subsets of the MB intrin-
sic neurons (Ferveur et al., 1995; O’Dell et al., 1995).
The MB’s role in mate discrimination was to be
reconsidered, however, because based on their large-
scale analysis of flies expressing traF in different
regions of the brain, Kido and Ito (2002) found that
bisexual behavior was not correlated with the femini-
zation of the MBs. Besides mate discrimination, the
involvement of the MBs in other aspects of experi-
ence-independent courtship behavior also still re-
mains elusive. According to Kido and Ito (2002), ab-
lation of the MBs by feeding newly hatched larvae
with hydroxyurea resulted in significant reduction in
the male courtship activity toward virgin females,
while the similar hydroxyurea treatment did not affect
the amount of their courtship in the study of McBride
et al. (1999). As for the CC, the functional study re-
garding experience-independent aspects of male
courtship is much limited. The involvement of the
CC in male courtship was indicated by the mutants
having structural defects in the CC. These mutants
were reported to start courtship with the same latent
period as wild-type flies, but to show significantly
lower copulation success rates, which could be attrib-
uted to their abnormal courtship song (Popov et al.,
2003).
Studies of flies with the partially feminized or
structurally damaged brain have provided useful in-
formation about central regulation of male courtship
behavior. Approaches using these flies, however, are
not sufficient to define the neural circuits for court-
ship, because the neurons playing a key role in court-
ship are not necessarily sexually dimorphic. In addi-
tion, irreversible developmental defects in particular
brain regions, induced chemically or genetically, may
cause functional abnormalities in other brain regions
that directly or indirectly interact with the damaged
neurons, which makes it difficult to attribute the
observed behavioral defects to the initially damaged
brain regions. To circumvent these potential prob-
lems, the temperature-sensitive Dynamin mutation
shibirets1 (shits1) can be used in combination with
the Gal4/UAS binary expression system to inhibit
neurotransmission in a spatially specific and tempe-
rature-dependent manner (Kitamoto, 2001, 2002).
Broughton et al. (2004) have recently utilized the
UAS-shits1 transgene with various Gal4 lines and ana-
lyzed the function of different brain regions in acute
physiological control of male courtship. They have
found that a cluster of cells in the lateral protocere-
Central Control of Male Courtship 823
brum (LPR), a higher-order neuropil region located
in the lateral part of the brain, plays a critical role in
courtship initiation.
Here we further expanded the usage of the UAS-
shits1 transgene and specifically focused on the acute
physiological functions of the MBs and the CC in
male courtship behavior. We found that blocking of
synaptic transmission in the MBs significantly af-
fected courtship initiation as well as maintenance
when virgin females were used as a sexual target. In-
terestingly, the same treatment did not change their
courtship behavior toward immature males that have
different pheromone profiles from virgin females and
elicit strong courtship behavior from sexually ma-
tured males (Tompkins et al., 1980; Hall, 1994). In con-
trast, blocking of synaptic transmission in a CC sub-
structure, the FB, reduces courtship activity slightly
but significantly toward both virgin females and
young males, with little effect on courtship initiation.
These results suggest that synaptic transmission in
the MBs plays a significant role in responding to
female-specific pheromones that stimulate courtship
initiation and maintenance, whereas neurotransmis-
sion in the FB is involved in the maintenance of
courtship regardless of the nature of courtship-stimu-
lating pheromones.
METHODS
Flies
A wild-type strain Canton-S (CS), eight enhancer-trap Gal4
lines, and a UAS-shits1 line were used in this study. OK348,
J183, c232, 30Y, OK107, and c772 were outcrossed for at
least five generations to white flies with Canton-S genetic
background. The origin of genetic background in c41 and
104Y is unknown. All flies were raised on glucose-yeast-
cornmeal medium at 23 6 18C in a 12:12 h light/dark (LD)
cycle (lights on at 8:00 am). Virgin males and females were
collected within 6 h after eclosion and maintained in vials
until experiments. All experiments were carried out during
daytime between 9:00 and 14:00 hours at 24.5 6 0.5 or 30
6 0.58C. The mating activity of D. melanogaster does not
change during the period between Zeitgeber Time (ZT) 0
and ZT 9 (Sakai and Ishida, 2001).
Microscopy
Adult male brains and appendages were dissected in PBS
and fixed with PBS containing 3.7% formaldehyde for
15 min at room temperature. GFP fluorescence was
observed with a confocal microscope (Zeiss 510). Z sec-
tions were collected at 2
m intervals and processed to con-
struct projections through an extended depth of focus.
Journal of Neurobiology. DOI 10.1002/neu
824 Sakai and Kitamoto
Recording and Analysis of
Male Courtship
A male and a target male or female were placed in an
acrylic plastic observation chamber (15 mm in diameter

3 mm in depth) using a manual aspirator. Courtship
behavior was video recorded (DCR-PC300; Sony, Japan)
and the courtship latency (CL) and courtship index (CI)
were determined. The CL was defined as the period from
the initial presentation of a target object to the initiation of
courtship behavior, and the CI was defined as the percent-
age of time the male spent performing courtship behavior in
a given observation period. We observed male courtship for
10 min or until the moment of copulation.
In the present study, we used three types of target ani-
mals: sexually mature CS females (3 days old), sexually
immature CS females (0–24 h old), and newly emerged
white males (0–4 h old). GAL4-line females were mated to
UAS-shits1 males and F1 male progeny (3–4 days old) were
used in the experiment. F1 males between GAL4-line
females and CS males (3–4 days old) or between CS
females and UAS-shits1 males (3–4 days old) were the con-
trol. We observed 32–56 males in each genotype and dis-
carded the data for males that did not show any courtship
behavior during the observation period. Experiments were
carried out either at 25 or 308C. For the 308C experiments,
experimental males were pretreated at the temperature for
30–50 min.
A courtship bout is defined as a male’s uninterrupted
display of any courtship elements. We measured frequency
of courtship bout and courtship bout duration by observing
courtship of 30 males in each genotype.
Measurement of the Locomotor Activity
The locomotor activity was measured with a Drosophila
Activity Monitor IV (TRIKINETICS, Waltham, MA). A
male fly (3–4 days old) was introduced into a glass tube
(4 mm in diameter and 65 mm in length) through which an
infrared beam crossed the path of the fly’s movement. The
activity monitor was shaken for 3 s and kept at 25 or 308C
for 30 min, followed by the measurement of the locomotor
activity for 20 min. The number of incidences at which the
fly crossed the infrared beam was counted every 1 min.
Statistics
In some cases values of the CL, CI, locomotor activity, and
courtship bout duration were not distributed normally.
Thus, we carried out the arcsine transformation of the data
of CI, and the log transformation of the data of the other pa-
rameter. When the transformed data distributed normally
and homoscedasticity was evident, such data were analyzed
using a two-sample t-test. When the data were not normally
distributed after the transformation, they were analyzed
using the Mann-Whitney U test. We used computer soft-
ware (SPSS 11.5J Base System for Windows) in these tests.
Journal of Neurobiology. DOI 10.1002/neu
RESULTS
Analysis of Gal4 Expression Patterns
Using the Green Fluorescent Protein
(GFP) Reporter Gene
In order to examine the physiological functions of the
MBs and the CC in male courtship, we used the tem-
perature-sensitive Dynamin mutation shits1 in combi-
nation with eight Gal4 lines that have been reported
to specifically mark the MBs or the CC. To confirm
and document expression patterns of these Gal4 lines,
we visualized Gal4-positive cells in the adult nervous
system by driving GFP.
Three Gal4 lines, OK107, 30Y, and c772, showed
strong GFP expression in the MBs as reported previ-
ously (Connolly et al., 1996; Zars et al., 2000;
McGuire et al., 2001) [Fig. 1(B–D)]. Besides the
MBs they showed scattered GFP expression in the
optic lobes [an asterisk in Fig. 1(B)] and restricted,
apparently nonoverlapping expression in the ALs
[arrowheads in Fig. 1(B)] as well as the SOG [an
arrow in Fig. 1(C)]. In these three lines GFP expres-
sion was also observed in the pars intercerebralis (PI)
located in the median part of the brain [arrow in
Fig. 1(B)]. Although weak, GFP expression was de-
tected in the EB within the CC for c772 (data not
shown).
We confirmed that three ‘‘FB’’ Gal4 lines, J183,
OK348, and 104Y (Joiner and Griffith, 1999; Con-
nolly et al., 1996; Rodan et al., 2002), directed GFP
expression in the FB, but not in the EB [Fig. 1(E–H)].
Expression outside the FB was most obvious in J183
[Fig. 1(F)], where the restricted subsets of the MB
neurons as well as scattered cells in the optic lobes
were GFP-positive [Fig. 1(E)]. OK348 is most spe-
cific to the FB among these three Gal4 lines [Fig.
1(G)], although faint expression was detected in the
MBs. In addition, a cluster of cells in the lateral
region of the brain showed prominent expression
[Fig. 1(G)]. In 104Y, GFP expression was also seen
in cells in the optic lobes and the SOG as well as
nerve bundles in the lateral regions of the brain [an
arrowhead in Fig. 1(H)]. GFP expression was not
observed in the MBs for this line [Fig. 1(H)].
GFP expression was highly restricted in the EB for
c232 [Fig. 1(I)] as reported by O’Dell et al. (1995).
c41 showed strong GFP-expression in the EB and weak
expression in subsets of the MB neurons [Fig. 1(J)].
GFP-positive cells were also seen in the PI as well as
the SOG for c41 [arrow and arrowhead in Fig. 1(J)].
We found that a few Gal4 lines showed GFP
expression in the limited groups of putative sensory
neurons associated with gustatory receptors in the
 Central Control of Male Courtship 825

Figure 2 Effects of temperature on male courtship behavior. The CL (A,C) and the CI (B,D)
were measured at 25 (open bars) and 308C (solid bars). Mature or immature virgin CS females were
used as a sexual target. We observed 24 CS males toward mature and immature females, 32 104Y/
þ males toward mature females, and 48 104Y/þ males toward immature females. We discarded
the data obtained from males that did not show courtship behavior during the observation period.
Error bars show SEM. *p < 0.05. NS, not significant. (A) Mean courtship latency in CS males. n ¼
23–24 in each bar. (B) Mean courtship index in CS males. n ¼ 23–24 in each bar. (C) Mean court-
ship latency in 104Y/þ males (104Y/þ; F1 progeny between 104Y females and CS males). n ¼
31–43 in each bar. (D) Mean courtship index in 104Y/þ males. n ¼ 31–43 in each bar.

proboscis and the leg (see Supplementary Material).
In labial palps of 30Y males, in addition to strong
GFP expression in non-neuronal structures called the
labellar glands, a subset of gustatory neurons showed
positive signals. GFP expression was also observed in
restricted putative sensory neurons in the tarsus of the
legs for 30Y, OK348, and J183 males. Apparently,
the sporadic PNS expression observed in these Gal4
lines did not overlap each other.
Effects of Disrupted Neurotransmission
in the MBs and the CC on Male Courtship
Behavior toward Virgin Females
To investigate the physiological functions of the
MBs and the CC in male courtship, we expressed
shits1 in these structures using the eight Gal4 lines
and examined the effects of conditional disruption
of Dynamin function on courtship activity toward
virgin females by comparing the relative numbers
for mean courtship latency and courtship index at
25 and 308C. Disruption of Dynamin function in
neurons most probably results in acute disruption of
neurotransmission. We used immature virgins, instead
of mature virgins, as a female target throughout this
study because under the conditions we used, the CIs
of control males (CS and 104Y/þ) toward mature vir-
gin females (3–4 days old) were slightly but signi-
ficantly higher at 30 than at 258C [Fig. 2(B,D)]. This
is apparently because mature virgins were more
receptive at 258C and copulated with males in shorter
time with less courtship than at 308C. The CI and
CL toward immature virgin females (0–24 h old),
which stimulate male courtship but do not copulate
(Gailey et al., 1984), were not affected by tempera-
ture change.
When shits1 expression was directed by OK107,
30Y, and c772 and temperature was raised, the male
flies showed a common phenotype in courtship to-
ward immature virgins. The latency to initiate court-
ship was significantly lengthened and the courtship
Journal of Neurobiology. DOI 10.1002/neu
826 Sakai and Kitamoto
Figure 3 Effects of shits1 expression in eight GAL4
lines on courtship behavior toward immature virgin
females at permissive and restrictive temperatures. Male
courtship behavior was observed at 25 (open bars) and
308C (solid bars). Error bars show SEM. NS, not signifi-
cant; *p < 0.05; **p < 0.01; ***p < 0.001. (A) Mean
courtship latency in three Gal4 lines (OK107, 30Y, and
c772) with strong expression in the MBs. n ¼ 32–40 in
each bar. (B) Mean courtship index in OK107, 30Y, and
c772. n ¼ 32–40 in each bar. (C) Mean courtship la-
tency in three Gal4 lines (J183, OK348, and 104Y) with
expression in the FB. n ¼ 35–56 in each bar. (D) Mean
courtship index in J183, OK348, and 104Y. n ¼ 35–56
in each bar. (E) Mean courtship latency in two Gal4
lines (c232 and c41) with expression in the EB. n ¼ 33–
40 in each bar. (F) Mean courtship index in c232 and
c41. n ¼ 33–40 in each bar.
Journal of Neurobiology. DOI 10.1002/neu
activity was reduced at 308C, compared to those at
258C [Fig. 3(A,B)]. Control males heterozygous for
either UAS-shits1 [CL at 258C ¼ 60.1 6 11.1 s (n
¼ 32), CL at 308C ¼ 70.0 6 13.3 s (n ¼ 26), Mann-
Whitney U ¼ 376, p ¼ 0.53; CI at 258C ¼ 85.1
6 2.2% (n ¼ 32), CI at 308C ¼ 81.6 6 2.6% (n ¼
26), Mann-Whitney U ¼ 327.5, p ¼ 0.17] or Gal4
insertions showed neither the lengthened CL nor
reduced CI at 308C [Fig. 3(A,B)]. Because OK107,
30Y, and c772 shared strong expression in the MBs,
but not in other brain regions or PNS neurons, the
results suggest that the disruption of synaptic trans-
mission in the MBs causes defects in courtship initia-
tion and maintenance. The reporter gene expression
was observed in PI neurons for all these lines. It is
not likely, however, that the PI is responsible for the
observed temperature-dependent phenotype, because
c41 with expression in the PI did not show the court-
ship defects [Fig. 3(E,F)].
Expression of shits1 using three Gal4 lines, J183,
OK348 and 104Y resulted in the same temperature-
dependent courtship phenotype. The CIs showed the
slight but statistically significant reduction at 30
compared to at 258C, whereas the CL was not
affected by the temperature change [Fig. 3(C,D)].
As shown in Figure 1 these Gal4 lines all showed
marked expression in the FB, suggesting that dis-
ruption of neurotransmission in the FB results in
reduction of courtship activity, but leaves courtship
initiation intact. J183 also showed reporter gene
expression in the MBs, but contrary to other MB
expressing Gal4 lines, the CL was not extended at
308C. The MB expression in J183 was relatively
weak and restricted to small subsets of the MB neu-
rons. It is possible that J183 does not direct gene
expression in the MB neurons that play a role in
courtship initiation. Alternatively, the expression
level of shits1 in the pertinent MB neurons was too
low to block synaptic transmission in J183/ UAS-
shits1. Similarly, OK348 showed only faint expres-
sion in the MBs, therefore the effect of shits1 in the
MBs might be disregarded.
In contrast to other Gal4 lines, c232/UAS-shits1
and c41/UAS-shits1 males did not show any temp-
erature-dependent abnormality in courtship [Fig. 3(E,
F)]. In these lines, distinct GAL4 expression was
seen in the EB but not in the FB (Fig. 1), implying
that disruption of the EB function does not affect
male courtship. Although c41 directed Gal4 expres-
sion in the MBs, the level of expression was very low
[Fig. 1(J)], which might explain why c41/UAS-shits1
males did not show defects in the CL and the CI as
seen in the Gal4 lines with marked expression in the
MBs.
 Central Control of Male Courtship 827

Figure 4 Effects of shits1 expression on courtship behavior toward newly emerged males at
permissive and restrictive temperatures. Male courtship behavior was observed at 25 (open bars)
and 308C (solid bars). Four Gal4 lines, c772, OK107, J183, and 104Y, were used in the exp-
eriments. Error bars show SEM. NS, not significant; *p < 0.05. (A) Courtship latency toward
immature males. n ¼ 30–39 in each bar. (B) Courtship index toward immature males. n ¼ 27–39 in
each bar.

Effects of Disrupted Neurotransmission UAS-shits1 and OK107/UAS-shits1 males. This indi-

in the MBs and the CC on Male Courtship
Behavior toward Immature Males
In Drosophila, newly emerged immature males have
different pheromone profiles from virgin females or
mature males, and elicit strong courtship behavior
from sexually matured males (Tompkins et al., 1980;
Hall, 1994). It is not well known whether courtship
behaviors toward virgin females and immature young
males require the same critical brain regions. We
therefore examined if male courtship toward imma-
ture males was affected when neurotransmission in
the MBs and the CC was disrupted.
In contrast to the earlier result obtained with virgin
females as a sexual target [Fig. 3(A,B)], courtship to-
ward immature males was not affected by tempera-
ture change for the males expressing shits1 in the MB
neurons using two independent Gal4 lines, c772 and
OK107. As shown in Figure 4(A) and (B), the levels
of the CL and the CI toward immature males did not
show significant differences at 25 and 308C for c772/
cates that the activity of the MBs neurons is not re-
quired for responding to courtship-stimulating cues re-
leased from immature males. On the other hand, the
temperature-dependent courtship phenotype toward
immature males was similar to that toward virgin
females when two lines (J183 and 104Y) driving shits1
expression in the FB were used. In these flies the
higher temperature slightly reduced the courtship ac-
tivity with no effect on courtship initiation (Fig. 4).
Although J183 expresses the reporter gene in the re-
stricted subsets of MB neurons [Fig. 1(E)], the charac-
teristic phenotype of J183/UAS-shits1 toward immature
males is not likely due to the inactivation of these neu-
rons because courtship toward immature males did not
show any temperature-dependent phenotype in c772
and OK107 directing extensive expression in the MBs.
This temperature-dependent courtship phenotype of
J183/UAS-shits1 and 104Y/UAS-shits1 is thus likely to
be due to the disruption of neurotransmission in the
FB, where J183 and 104Y share expression.
Journal of Neurobiology. DOI 10.1002/neu
828 Sakai and Kitamoto
Effects of Disrupted Neurotransmission
in the MBs and the CC on Courtship Bout
Duration
In observations of courtship performed by Gal4 lines
expressing shits1 in the MBs or the CC, we noticed
that it was frequently interrupted and did not last
long. To quantify this defect in courtship, we re-
corded the duration of each courtship bout performed
by 30 males in each genotype at 25 and 308C. The
mean duration of courtship bout was calculated from
total courtship bouts observed for this analysis.
When neurotransmission in the MBs was disrupted
at 308C in c772/UAS-shits1 males, the frequency of
short courtship bouts (<60 s) toward virgin females
was strikingly increased and the mean duration of
courtship bouts was greatly reduced [Fig. 5(A)]. In
contrast, temperatures did not affect mean duration of
courtship bouts toward newly emerged males in
c772/UAS-shits1 [Fig. 5(B)], showing that the effect
is specific to courtship toward virgin females. In both
c772/þ and þ/UAS-shits1 control males, temperature
did not significantly affect the mean duration of
courtship bouts regardless of the sexual targets [Fig.
5(A,B)]. These results indicate that the on-going ac-
tivity of the MB neurons is required for maintaining
courtship activity when virgin females are used as a
sexual target. J183/UAS-shits1 males showed the
reduced mean duration of courtship bouts at 308C
both toward virgin females and newly emerged
immature males [Fig. 5(C,D)], which resulted in
reduction of the overall courtship activity regardless
of the sexual targets [Figs. 3(D) and 4(B)].
Locomotor Activity at Permissive and
Restrictive Temperatures
The general locomotor activity in flies plays an im-
portant role in male courtship behavior. Kyriacou
(1981) reported that high locomotor activity levels
were positively correlated with superiority in male
mating performance in D. melanogaster. It is possible
that the courtship defects observed in males with
functionally disrupted MB or FB neurons are simply
due to their low locomotor activity. To address this
issue we measured their locomotor activity at 25 and
308C every 1 min for 20 min (see Methods).
The locomotor activity was higher at 30 than at
258C in all control males (Gal4/þ and þ/UAS-shits1)
except for c41/þ and 30Y/þ (Fig. 6, solid lines or
bars). When shits1 expression was directed to the
MBs using OK107 and c772, the temperature-
induced increase in the locomotor activity was further
enhanced in comparison with Gal4/þ and þ/UAS-
Journal of Neurobiology. DOI 10.1002/neu
shits1 controls [Fig. 6(A,B)], indicating that the activ-
ity of the MB neurons normally suppresses general
locomotor activity. This result is consistent with the
report by Martin et al. (1998) demonstrating that
chemical ablation of the MBs, a mutation affecting
Kenyon cell differentiation (mushroom body minia-
ture1), and the disrupted MB neurotransmission all
induced the elevated locomotion. As for the Gal4
lines directing expression in the FB, OK348/UAS-
shits1 males showed the higher locomotor activity at
30 than at 258C, whereas the levels of the locomotor
activity at 30 and 258C were similar in J183/UAS-
shits1 and 104Y/UAS-shits1 males [Fig. 6(C,D)]. In
c232 and c41, disruption of neurotransmission in the
EB did not affect locomotor activity [Fig. 6(E,F)]. In
any case, the locomotor activity at 308C was not
lower than that measured at 258C for all GAL4 lines
expressing shits1. Therefore, it is not likely that the
reduced locomotor activity is a cause of temperature-
dependent courtship defects observed in Gal4 lines
expressing shits1 in the MBs or the FB.
DISCUSSION
In this study we focused on two major structures in
the Drosophila brain, the MBs and the CC, and exam-
ined their physiological roles in experience-independ-
ent aspects of male courtship behavior using the shits1
transgene in combination with various Gal4 lines.
Since GAL4 expression is not absolutely limited to
the MB or CC in ‘‘MB lines’’ or ‘‘CC lines’’ used in
this study, we cannot formally rule out the possibility
that neurons outside the MB or CC somehow contrib-
ute to their shits1-dependent courtship phenotypes.
However, it is most likely that suppression of MB or
CC functions is the main cause of the observed court-
ship phenotypes, because: (1) these GAL4 lines show
strong, overlapping GAL4 expression in the MB or
CC; (2) all ‘‘MB lines’’ or ‘‘CC lines’’ show, in
essence, the same tendency in shits1 -dependent court-
ship phenotypes; (3) there are no GAL4-positive
structures in the brain that are common to the all
‘‘MB lines’’ or ‘‘CC lines’’ other than the MB or CC
(except for the PI neurons in the MB lines); and (4)
involvement of the MB and CC in male courtship
was suggested in other studies.
We found that the acute disruption of Dynamin
function using the shits1 transgene, most probably
resulting in blockage of synaptic neurotransmission,
in the MBs leads to two clear defects in male court-
ship toward virgin females [Fig. 3(A,B)]. First, the
length of the period from the initial presentation of a
virgin female to the initiation of courtship behavior is
 Central Control of Male Courtship 829

Figure 5 Effects of shits1 expression in c772 and J183 on courtship bout duration toward imma-
ture virgin females and newly emerged males at permissive and restrictive temperatures. NS, not
significant; *p < 0.05; ***p < 0.001. Gal4/þ and þ/UAS-shits1 males were used as a control. (A)
Histograms and mean duration of courtship bout toward virgin females in c772. (B) Histograms
and mean duration of courtship bout toward newly emerged males in c772. (C) Histograms and
mean duration of courtship bout toward virgin females in J183. (D) Histograms and mean duration
of courtship bout toward newly emerged males in J183.

Journal of Neurobiology. DOI 10.1002/neu

 Figure 6

Journal of Neurobiology. DOI 10.1002/neu

 Central Control of Male Courtship 831

significantly extended (initiation defect). Second,
courtship behavior is often interrupted and therefore
the length of each courtship bout is shortened (main-
tenance defect). Interestingly, these defects are not
apparent in courtship toward immature males (Fig. 4).
Multiple sensory modalities are integrated to trigger
male courtship behavior. Loss of one of the inputs
does not abolish courtship, but reduces the intensity
and quality of the behavior. Inactivation of the MB
neurons is likely to deprive males, at least partially,
of the chemical information from virgin females. Our
finding suggests that MB-mediated information proc-
essing of chemical cues from virgin females plays an
important role in courtship initiation as well as main-
tenance. Although the MBs are not essential for
males to accomplish copulation in laboratory court-
ship settings, it is possible that male flies with defec-
tive MBs would be much less fertile in nature
because of their subnormal courtship behavior.
In support of our finding, Broughton et al. (2004)
reported that the courtship activity toward virgin
females was reduced when synaptic transmission was
disrupted using UAS-shits1 and the Gal4 lines with
strong expression in the MBs. In Broughton et al.
(2004), however, the courtship initiation defect was
not detected in males with disrupted MB neurotrans-
mission. This apparent discrepancy could be ex-
plained by the different experimental conditions between
the two studies. The chamber used for assaying court-
ship in the current study was larger than the one in
Broughton et al. (2004) (15 mm in diameter
3 mm
in depth vs. 8
3 mm), resulting in relatively longer
courtship latency. In a larger chamber, courtship ini-
tiation might be more dependent on the sensory infor-
mation processed by the MB neurons, therefore the
effect of MB dysfunctions could be more easily
detected. In addition, male flies were pretreated at
308C for a longer period in this study (30–50 min vs.
10 min), which may lead to more severe inhibition of
the MB neurotransmission.
A short pulse of hydroxyurea administration to
newly hatched larvae results in ablation of the MBs
(de Belle and Heisenberg, 1994). As mentioned in the
Introduction, the MB ablation by the hydroxyurea
treatment was reported to cause a significant decrease
in the activity of male courtship in one study (Kido
and Ito, 2002), while in the other study the MB abla-
tion was reported to have no effect on the quality and
quantity of male courtship (McBride et al., 1999).
Although it is not known why the similar chemical
treatment leads to different behavioral consequences,
the possibility exists that the extent of structural
defects in the MBs might be somehow different
between the two studies. Our observation that block-
ing of synaptic transmission in the MB intrinsic neu-
rons delays courtship initiation and reduces the court-
ship activity rather fits in the result presented by Kido
and Ito (2002). On the other hand, a possibility cannot
be ruled out at least formally that non-MB neurons
could compensate for a role of the MBs in courtship
when MB neurons are removed irreversibly during
development. If that would be the case, our data
would not contradict the result presented by McBride
et al. (1999).
Heimbeck et al. (2001) found that male courtship
was severely impaired when synaptic transmission
from the two-thirds of all projection neurons (PNs)
was blocked by ectopic expression of tetanus toxin
light chain. PNs are the major elements of the central
olfactory pathway, which receive information from
olfactory receptor neurons in the first-order olfactory
neuropils, the ALs (Stocker et al., 1990; Stocker,
1994). PNs transmit the information from the ALs to
a higher order neuropil region, the LPR, through two
different pathways, either directly without passing
through the MBs or indirectly via the MBs (Stocker
et al., 1990; Stoker, 1994). Heimbeck et al. (2001)
have concluded that the direct neuronal pathway from
ALs to the LPR, which is not mediated by the MBs,
is necessary and sufficient to process male courtship.
However, the argument was based on the assumption
that the MBs are not involved in experience-inde-
pendent courtship (McBride et al., 1999). Our results
indicate that the indirect neuronal pathway from ALs

Figure 6 Effects of shits1 expression in eight GAL4 lines on locomotor activity at permissive and
restrictive temperatures. The locomotor activity was measured for 20 min at 25 (dotted lines or
open bars) and 308C (solid lines or bars). Mean locomotor activity every 1 min and mean total loco-
motor activity for 20 min were calculated in each genotype. Gal4/þ and þ/UAS-shits1 males were
used as a control. Error bars show SEM. NS, not significant; **p < 0.01; ***p < 0.001. (A) Mean
locomotor activity every 1 min in UAS-shits1, OK107, 30Y, and c772. n ¼ 42–48. (B) Mean total
locomotor activity in UAS-shits1, OK107, 30Y, and c772. n ¼ 42–48. (C) Mean locomotor activity
every 1 min in J183, OK348, and 104Y. n ¼ 41–48. (D) Mean total locomotor activity in J183,
OK348, and 104Y. n ¼ 41–48. (E) Mean locomotor activity every 1 min in c232 and c41. n ¼ 40–
48. (F) Mean total locomotor activity in c232 and c41. n ¼ 40–48.

Journal of Neurobiology. DOI 10.1002/neu

832 Sakai and Kitamoto
to the LPR through the MBs also participates in
courtship regulation.
The fruitless gene (fru) plays a critical role in male
courtship behavior. Different fru mutant alleles and
their combinations variously disrupt each step of
male courtship (Gailey and Hall, 1989; Villella et al.,
1997). Recently Demir and Dickson (2005) and Man-
oli et al. (2005) have shown that male-specific fru
protein (FruM), which is expressed in 2% of neu-
rons in the male nervous system, is both necessary
and sufficient to build the potential for male courtship
behavior into the Drosophila nervous system. Stock-
inger et al. (2005) and Manoli et al. (2005) have tar-
geted the insertion of Gal4 into the fru locus and
visualized the neurons expressing male-specific fru
protein (FruM) and their projections. Many different
synaptic neuropils in the brain are labeled for FruM,
including the /-lobe and -lobe of the MBs.
Because most, if not all, FruM-expressing neurons are
expected to be involved in male sexual behavior, this
is another piece of data indicating the involvement of
the MBs in courtship.
Compared to the functional disruption of the MBs,
that of the CC had different effects on male courtship
(Figs. 3–5). Of distinct neuropil regions in the CC, our
results indicate that the FB, but not the EB, partici-
pates in the proper maintenance of male courtship.
The CC has been proposed as a principal site of loco-
motor control (Strauss, 2002). Supporting this pro-
posal, Martin et al. (1999) showed that continuous
blocking of the synaptic transmission by the targeted
expression of tetanus toxin in the FB or the EB indu-
ces reduction of locomotor activity. However, the
slight reduction of the courtship activity observed in
male flies with acutely impaired FB neurotransmission
cannot be simply explained by a change in locomotor
activity, because we could not detect significant
reduction in general locomotor activity under the con-
ditions where courtship behavior was recorded (Fig. 6).
It is not known why the acute disruption of synaptic
transmission in the FB leads to reduction of the male
courtship activity. One possibility is that the fine reg-
ulation of motor outputs, other than the general loco-
motor activity, is affected by the disruption of FB
neurotransmission, which contributes to the frequent
termination of courtship behavior. Alternatively, it is
also possible that courtship is defective in the FB-
impaired males because the FB neurons integrate sen-
sory information and play a role in keeping courtship
motivation or interest at a certain state. The reduction
of the courtship activity in the FB-impaired males
was observed for different sexual targets that release
different courtship-stimulating pheromones, implying
that the FB controls courtship motivation or interest
Journal of Neurobiology. DOI 10.1002/neu
at a higher level. In this regard, it should be pointed
out that CaMKII activity in the FB is required for the
experience-dependent modification of courtship
behavior called ‘‘courtship conditioning’’ or ‘‘condi-
tioned courtship suppression’’, where a male fly dis-
plays a reduced amount of courtship toward virgin
females after being paired with a previously mated
female (Siegel and Hall, 1979; Joiner and Griffith,
1999). Courtship conditioning is considered to repre-
sent an association of an attractive pheromone released
by both virgin and mated females and an aversive
pheromone released only by mated females (Tomp-
kins et al., 1983). An interesting hypothesis is that ex-
perience with a previously mated female causes stable
physiological changes in the FB of the conditioned
male, which in turn leads to the reduction of courtship
motivation or interest that lasts for hours and days.
Recently we have reported that a circadian clock gene,
period, plays a key role in the formation of long-term
courtship memory (Sakai et al., 2004). Because the
period gene is expressed in the FB (Kaneko and Hall,
2000), per functions might be involved in the possi-
ble changes induced in the FB after a male has an ex-
perience with a mated female. The synaptic transmis-
sion in the FB may play a role not only in experience-
independent aspects of courtship but also in its expe-
rience-dependent modification.
We thank L. C. Griffith, U. Heberlein, T. Zars, R. L.
Davis, M. Saitoe, and J. D. Armstrong for providing GAL4
lines, and M. A. Joiner for useful discussion.
REFERENCES
Baker BS, Taylor BJ, Hall JC. 2001. Are complex behav-
iors specified by dedicated regulatory genes? Reasoning
from Drosophila. Cell 105:13–24.
Bastock M, Manning A. 1955. The courtship of Drosophila
melanogaster. Behaviour 15:85–111.
Besson M, Martin JR. 2005. Centrophobism/thigmotaxis, a
new role for the mushroom bodies in Drosophila. J Neu-
robiol 62:386–396.
Bray S, Amrein H. 2003. A putative Drosophila pheromone
receptor expressed in male-specific taste neurons is re-
quired for efficient courtship. Neuron 39:881–883.
Broughton SJ, Kitamoto T, Greenspan RJ. 2004. Excitatory
and inhibitory switches for courtship in the brain of Dro-
sophila melanogaster. Curr Biol 14:538–547.
Cobb M, Burnet B, Connolly K. 1986. The structure of
courtship in the Drosophila melanogaster species sub-
group. Behaviour 97:182–212.
Connolly JB, Roberts IJ, Armstrong JD, Kaiser K, Forte M,
Tully T, O’Kane CJ. 1996. Associative learning dis-
rupted by impaired Gs signaling in Drosophila mush-
room bodies. Science 274:2104–2107.

De Belle JS, Heisenberg M. 1994. Associative odor learn-
ing in Drosophila abolished by chemical ablation of
mushroom bodies. Science 263:692–695.
Demir E, Dickson BJ. 2005. fruitless splicing specifies male
courtship behavior in Drosophila. Cell 121:785–794.
Dubnau J, Grady L, Kitamoto T, Tully T. 2001. Disruption
of neurotransmission in Drosophila mushroom body
blocks retrieval but not acquisition of memory. Nature
411:476–480.
Ferveur JF, Jallon JM. 1996. Genetic control of male cutic-
ular hydrocarbon in Drosophila melanogaster. Genet Res
67:211–218.
Ferveur JF, Stortkuhl KF, Stocker RF, Greenspan RJ. 1995.
Genetic feminization of brain structures and changed
sexual orientation in male Drosophila. Science 267:902–
905.
Gailey DA, Hall JC. 1989. Behavior and cytogenetics of
fruitless in Drosophila melanogaster: different courtship
defects caused by separate, closely linked lesions. Ge-
netics 121:773–785.
Gailey DA, Jackson FR, Siegel RW. 1984. Conditioning
mutations in Drosophila melanogaster affect an expe-
rience-dependent behavioral modification in courting
males. Genetics 106:613–623.
Greenspan RJ, Ferveur JF. 2000. Courtship in Drosophila.
Annu Rev Genet 34:205–232.
Hall JC. 1979. Control of male reproductive behavior by the
central nervous system of Drosophila: dissection of a court-
ship pathway by genetic mosaics. Genetics 92:437–457.
Hall JC. 1994. The mating of a fly. Science 264:1702–
1714.
Hanesch U, Fischbach KF, Heisenberg M. 1989. Neuronal
architecture of the central complex in Drosophila mela-
nogaster. Cell Tissue Res 257:343–366.
Heimbeck G, Bugnon V, Gendre N, Keller A, Stocker RF.
2001. A central neural circuit for experience-independent
olfactory and courtship behavior in Drosophila mela-
nogaster. Proc Natl Acad Sci USA 98:15336–15341.
Heisenberg M, Borst A, Wagner S, Byers D. 1985. Dro-
sophila mushroom body mutants are deficient in olfac-
tory learning. J Neurogenet 2:1–30.
Hotta Y, Benzer S. 1976. Courtship in Drosophila mosaics:
Sex-specific foci for sequential action patterns. Proc Natl
Acad Sci USA 73:4154–4158.
Ilius M, Wolf R, Heisenberg M. 1994. The central complex
of Drosophila melanogaster is involved in flight control:
studies on mutants and mosaics of the gene ellipsoid
body open. J Neurogenet 9:189–206.
Ito K, Suzuki K, Estes P, Ramaswami M, Yamamoto D,
Strausfeld NJ. 1998. The organization of extrinsic neu-
rons and their implications in the functional roles of the
mushroom bodies in Drosophila melanogaster Meigen.
Learn Mem 5:52–77.
Jallon JM. 1984. A few chemical words exchange by Dro-
sophila during courtship and mating. Behav Genet 14:
441–478.
Joiner MA, Griffith LC. 1999. Mapping of the anatomical
circuit of CaM kinase-dependent courtship conditioning
in Drosophila. Learn Mem 6:177–192.
Central Control of Male Courtship 833
Kaneko M, Hall JC. 2000. Neuroanatomy of cells express-
ing clock genes in Drosophila: transgenic manipulation
of the period and timeless genes to mark the perikarya
of circadian pacemaker neurons and their projections.
JComp Neurol 422:66–94.
Kido A, Ito K. 2002. Mushroom bodies are not required for
courtship behavior by normal and sexually mosaic Dro-
sophila. J Neurobiol 52:302–311.
Kitamoto T. 2001. Conditional modification of behavior in
Drosophila by targeted expression of a temperature-sen-
sitive shibire allele in defined neurons. J Neurobiol 47:
81–92.
Kitamoto T. 2002. Targeted expression of temperature-
sensitive dynamin to study neural mechanisms of com-
plex behavior in Drosophila. J Neurogenet 16:205–
228.
Kyriacou CP. 1981. The relationship between locomotor
activity and sexual behaviour in ebony strains of Dro-
sophila melanogaster. Anim Behav 29:462–471.
Liu L, Wolf R, Ernst R, Heisenberg M. 1999. Context gen-
eralization in Drosophila visual learning requires the
mushroom bodies. Nature 400:753–756.
Manoli DS, Foss M, Villella A, Taylor BJ, Hall JC, Baker
BS. 2005. Male-specific fruitless specifies the neural sub-
strates of Drosophila courtship behaviour. Nature 436:
395–400.
Markow TA, Manning M. 1980. Mating success of photore-
ceptor mutants of Drosophila melanogaster. Behav Neur
Biol 29:276–280.
Martin JR, Ernst R, Heisenberg M. 1998. Mushroom bodies
suppress locomotor activity in Drosophila melanogaster.
Learn Mem 5:179–191.
Martin JR, Raabe T, Heisenberg M. 1999. Central complex
substructures are required for the maintenance of loco-
motor activity in Drosophila melanogaster. J Comp Phys-
iol [A] 185:277–288.
McBride SM, Giuliani G, Choi C, Krause P, Correale D,
Watson K, Baker G, et al. 1999. Mushroom body abla-
tion impairs short-term memory and long-term memory
of courtship conditioning in Drosophila melanogaster.
Neuron 24:967–977.
McGuire SE, Le PT, Davis RL. 2001. The role of Drosoph-
ila mushroom body signaling in olfactory memory. Sci-
ence 293:1330–1333.
O’Dell KM, Armstrong JD, Yang MY, Kaiser K. 1995.
Functional dissection of the Drosophila mushroom
bodies by selective feminization of genetically defined
subcompartments. Neuron 15:55–61.
Popov AV, Sitnik NA, Savvateeva-Popova EV, Wolf R,
Heisenberg M. 2003. The role of central parts of the
brain in the control of sound production during courtship
in Drosophila melanogaster. Neurosci Behav Physiol
33:53–65.
Rodan AR, Kiger JA Jr, Heberlein U. 2002. Functional dis-
section of neuroanatomical loci regulating ethanol sensi-
tivity in Drosophila. J Neurosci 22:9490–9501.
Sakai T, Ishida N. 2001. Circadian rhythms of female mat-
ing activity governed by clock genes in Drosophila. Proc
Natl Acad Sci USA 98:9221–9225.
Journal of Neurobiology. DOI 10.1002/neu
834 Sakai and Kitamoto
Sakai T, Isono K, Tomaru M, Fukatami A, Oguma Y. 2002.
Light wavelength dependency of mating activity in the
Drosophila melanogaster species subgroup. Genes Genet
Syst 77:187–195.
Sakai T, Isono K, Tomaru M, Oguma Y. 1997. Light-affected
male following behavior is involved in light-dependent
mating in Drosophila melanogaster. Genes Genet Syst 72:
275–281.
Sakai T, Tamura T, Kitamoto T, Kidokoro Y. 2004. A clock
gene, period, plays a key role in long-term memory for-
mation in Drosophila. Proc Natl Acad Sci USA 101:
16058–16063.
Siegel RW, Hall JC. 1979. Conditioned courtship responses
in courtship behavior of normal and mutant Drosophila.
Proc Natl Acad Sci USA 76:3430–3434.
Stocker RF. 1994. The organization of the chemosensory
system in Drosophila melanogaster: a review. Cell Tissue
Res 275:3–26.
Stocker RF, Lienhard MC, Borst A, Fischbach KF. 1990.
Neuronal architecture of the antennal lobe in Drosophila
melanogaster. Cell Tissue Res 262:9–34.
Stockinger P, Kvitsiani D, Rotkopf S, Tirian L, Dickson BJ.
2005. Neural circuitry that governs Drosophila male
courtship behavior. Cell 121:795–807.
Strausfeld NJ, Hansen L, Li Y, Gomez RS, Ito K. 1998.
Evolution, discovery, and interpretations of arthropod
mushroom bodies. Learn Mem 5:11–32.
Journal of Neurobiology. DOI 10.1002/neu
Strauss R. 2002. The central complex and the genetic dis-
section of locomotor behavior. Curr Opin Neurobiol 12:
633–638.
Strauss R, Heisenberg M. 1993. A higher control center of
locomotor behavior in the Drosophila brain. J Neurosci
13:1852–1861.
Tompkins L. 1984. Genetic analysis of sex appeal in Dro-
sophila. Behav Genet 14:411–440.
Tompkins L, Hall JC, Hall ML. 1980. Courtship-stimulat-
ing volatile compounds from normal and mutant Dro-
sophila. J Insect Physiol 26:689–697.
Tompkins L, Siegel RW, Gailey DA, Hall JC. 1983.
Conditioned courtship in Drosophila and its mediation
by association of chemical cues. Behav Genet 13:565–
578.
Villella A, Gailey DA, Berwald B, Ohshima S, Barnes PT,
Hall JC. 1997. Extended reproductive roles of the fruit-
less gene in Drosophila melanogaster revealed by behav-
ioral analysis of new fru mutants. Genetics 147:1107–
1130.
Welbergen P, van Dijken FR, Scharloo W. 1987. Collation
of the courtship behaviour of the sympatric species Dro-
sophila melanogaster and Drosophila simulans. Behav-
iour 101:253–274.
Zars T, Fischer M, Schulz R, Heisenberg M. 2000. Local-
ization of a short-term memory in Drosophila. Science
288:672–675.